Across 23 comparable benchmarks, Nemotron 3.5 Super VL has better reported results on 12 and worse on 11. 2 further shared results are listed below, each with the reason it is not counted.
Moderate evidence23 of 25 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Nemotron 3.5 Super VL | Muse Glimmer-30B |
|---|---|---|
| Context window | 1M tokens | 128K tokens |
| Input / 1M | $0.150 | $0.300 |
| Output / 1M | $0.500 | $1.20 |
| Blended (4:1) | $0.126 | $0.293 |
| Status | Available | Available |
| Released | October 15, 2026 | August 10, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 15 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Nemotron 3.5 Super VL ahead on 3 of 4 counted
Nemotron 3.5 Super VL ahead on 1 of 1 counted
Muse Glimmer-30B ahead on 3 of 3 counted
Nemotron 3.5 Super VL ahead on 2 of 2 counted
Split 1–1 of 2 counted
Muse Glimmer-30B ahead on 4 of 7 counted
Split 1–1 of 2 counted
No fixed-scale average — see the rows in Benchmark results
No shared comparable results for Abstract reasoning, Mathematics, Cybersecurity.
Published scores, with matching modes where available. Gaps show Nemotron 3.5 Super VL relative to Muse Glimmer-30B. Reading the results
| Benchmark | Nemotron 3.5 Super VL | Muse Glimmer-30B | Gap |
|---|---|---|---|
| DeepSWERepository Engineering · Pass@1 (DeepSWE v1.1) | 24.6Unspecified Reasoning · tools | 4.4≠Thinking · tools | +20.3 pts#69 vs #72 of 73 |
| FrontierCode 2.0Repository Engineering · Weighted rubric score | 6.1Thinking · tools | 1.5Thinking · tools | 4.07×#32 vs #35 of 36 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 71.1Unspecified Reasoning · tools | 76.0≠Thinking High · tools | −4.9 pts#56 vs #34 of 108 |
| Terminal-Bench 2.1Agentic Development · Accuracy | 60.7Unspecified Reasoning · tools | 51.7≠Thinking High · tools | +9.0 pts#59 vs #67 of 68 |
| GPQA DiamondScientific Reasoning · Accuracy | 89.5Reported best | 83.5≠Thinking High · no tools | +6.0 pts#54 vs #92 of 200 |
| MCP-AtlasTool Orchestration · Pass rate / claim coverage | 67.7Thinking · tools | 75.5≠Thinking High · tools | −7.8 pts#41 vs #31 of 55 |
| OSWorld-VerifiedDesktop Workflows · Accuracy | 51.1Unspecified Reasoning · tools | 65.9≠Thinking High · tools | −14.8 pts#30 vs #25 of 36 |
| ScreenSpot-Pro (no tools)GUI Grounding · Success rate (%) | 43.5Thinking · no tools | 75.4Thinking · no tools | −31.9 pts#7 vs #5 of 8 |
| BrowseCompFact Finding · Accuracy | 78.0Unspecified Reasoning · tools | 63.9≠Thinking · tools | +14.1 pts#26 vs #39 of 61 |
| DeepSearchQAFact Finding · F1 score | 77.7Thinking · tools | 74.6≠Thinking High · tools + web | 1.04×#7 vs #8 of 9 |
| AA-LCRLong Reasoning · Accuracy | 68.6Reported best | 80.0≠Thinking High · no tools | −11.4 pts#18 vs #14 of 22 |
| LongBench v2Long Reasoning · Accuracy | 62.1Thinking · no tools | 33.1Thinking · no tools | +29.0 pts#7 vs #22 of 22 |
| ChartographyDocuments & Charts · Score | 29.3Thinking · tools | 25.1Thinking · tools | 1.17×#26 vs #28 of 29 |
| CharXiv RQDocuments & Charts · Score | 80.8Reported best | 78.8≠Thinking High · no tools | 1.03×#18 vs #19 of 21 |
| GDP.pdfDocuments & Charts · Task pass rate (%) | 8.0Thinking · no tools | 11.8Thinking · no tools | −3.8 pts#66 vs #54 of 96 |
| MMMU-ProVisual Understanding · Accuracy | 72.3Thinking · no tools | 74.0≠Thinking High · no tools | −1.7 pts#81 vs #72 of 136 |
| OmniDocBenchDocuments & Charts · Accuracy | 77.9Thinking · no tools | 75.8≠Thinking High · no tools | +2.1 pts#5 vs #6 of 6 |
| PerceptionBenchVisual Understanding · Accuracy | 36.8Thinking · no tools | 42.8Thinking · no tools | −6.0 pts#19 vs #17 of 19 |
| ZeroBench MainVisual Understanding · Pass@1 / Pass@5 / Pass^5 | 12.0Thinking · tools | 17.0Thinking · tools | −5.0 pts#19 vs #18 of 21 |
| GDPval-AA v2Cross-industry Work · Elo score | 1211Unspecified Reasoning · tools | 953≠Thinking High · tools | 1.27×#20 vs #21 of 23 |
| JobBenchCross-industry Work · Mean task reward (%)Not counted · This metric has no better direction | 36.5Thinking · tools | 16.1Thinking · tools | — |
| τ³-BankingService Workflows · Score | 18.9Unspecified Reasoning · tools | 23.5≠Thinking High · tools | 0.80×#60 vs #50 of 120 |
| HLEKnowledge Exams · AccuracyNot counted · Different or unrecorded test environments — excluded from the tally | 31.8Reported best· text-only | 22.0≠Thinking High · no tools· legacy-default | —#85 vs #95 of 151 |
| HLE-DiamondKnowledge Exams · Accuracy (%) | 8.8Thinking · no tools | 9.1Thinking · no tools | −0.3 pts#15 vs #14 of 16 |
| AA Intelligence Index v5.0.1Capability Indices · Index (0-100) | 13.6Thinking High · tools | 6.3Thinking High · tools | 2.15×#49 vs #55 of 62 |
Repository Engineering · Pass@1 (DeepSWE v1.1)
Repository Engineering · Weighted rubric score
Repository Engineering · Accuracy
Agentic Development · Accuracy
Scientific Reasoning · Accuracy
Tool Orchestration · Pass rate / claim coverage
Desktop Workflows · Accuracy
GUI Grounding · Success rate (%)
Fact Finding · Accuracy
Fact Finding · F1 score
Long Reasoning · Accuracy
Long Reasoning · Accuracy
Documents & Charts · Score
Documents & Charts · Score
Documents & Charts · Task pass rate (%)
Visual Understanding · Accuracy
Documents & Charts · Accuracy
Visual Understanding · Accuracy
Visual Understanding · Pass@1 / Pass@5 / Pass^5
Cross-industry Work · Elo score
Cross-industry Work · Mean task reward (%)
Not counted · This metric has no better direction
Service Workflows · Score
Knowledge Exams · Accuracy
Not counted · Different or unrecorded test environments — excluded from the tally
Knowledge Exams · Accuracy (%)
Capability Indices · Index (0-100)
A dash means no result is available.
| Benchmark | Nemotron 3.5 Super VL | Muse Glimmer-30B |
|---|---|---|
| AIME 2026 | — | 94.7 |
| CritPt | 10.9 | — |
| Riemann-bench | — | 13.6 |
| SciCode | — | 43.6 |
| SWE-bench Multilingual | 67.7 | — |
| SWE-Bench Pro - Public | — | 51.2 |
| Terminal-Bench 4.0 | 0.0 | — |
| Terminal-Bench 4.1 | 0.3 | — |
| Vibe Code Bench v1.1 | 3.5 | — |
| AutomationBench-AA | — | 6.8 |
| LiveBench 2.0 | 56.0 | — |
| Vals Index v2.2 | 29.8 | — |
| Creative Writing | — | 1804 |
| IF Bench | — | 77.0 |
| OpenAI MRCR v2 (8-needle, 256K–512K) | 5.7 | — |
| BabyVision | 29.6 | — |
| MathVision | 78.2 | — |
| Video-MME v2 | 32.9 | — |
| AA-AnalystAgent | — | 20.0 |
| AA-Briefcase | — | 478 |
| τ²-Bench - Telecom | 73.2 | — |
| ARC-AGI-1 | 22.7 | — |
| ARC-AGI-2 | 1.9 | — |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
Nemotron 3.5 Super VL costs57%lessthan Muse Glimmer-30B per million tokens — 2.3× cheaper.Using standard rates at this token mix.
Nemotron 3.5 Super VL: Standard pricing. DeepInfra official models API (nvidia/NVIDIA-Nemotron-3.5-Super-VL) verified 2026-11-13
Muse Glimmer-30B: Standard pricing. DeepInfra official models API (meta-models/Muse-Glimmer-30B) verified 2026-11-13
| Price / 1M tokens | Nemotron 3.5 Super VL | Muse Glimmer-30B | Ratio |
|---|---|---|---|
| Input | $0.15 | $0.30 | 0.50× |
| Output | $0.50 | $1.20 | 0.42× |
| Cached input | $0.02 | $0.04 | 0.50× |
| Cache write | — | — | — |
| Blended | $0.126 | $0.293 | 0.43× |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
Nemotron 3.5 Super VL relative to Muse Glimmer-30B
Summary: 1 of 1 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 1 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Nemotron 3.5 Super VL 3.1× more expensive
Nemotron 3.5 Super VL 3.1× more expensive
Effort, protocol, aggregation and source match; other run conditions may differ.
Nemotron 3.5 Super VL
Muse Glimmer-30B
| Benchmark | Nemotron 3.5 Super VL | Muse Glimmer-30B |
|---|---|---|
| ARC-AGI-1 | Not reported | |
| ARC-AGI-2 | Not reported | |
| LiveBench 2.0 | $0.1577 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 56.0 Max (task-specific tools) | Not reported |
| Terminal-Bench 4.0 | $0.86 / task Average · Resolution rate (%): 0.0 Unspecified Reasoning (With Tools) · vals-mini-swe | Not reported |
| Terminal-Bench 4.1 | Not reported |
| Attribute | Nemotron 3.5 Super VL | Muse Glimmer-30B |
|---|---|---|
| Availability | Available | Available |
| Provider | NVIDIA | Facebook AI Research Lab |
| Model type | Multimodal model | Reasoning model |
| First release | October 15, 2026 | August 10, 2026 |
| Context window | 1M tokens | 128K tokens |
| Total parameters | 120B | 29.6B |
| Active parameters | 12B | 29.6B |
| Architecture | Mixture of Experts (MoE) | Dense |
| Documented capabilities | Reasoning, Multilingual, Tool use, Function calling, Computer use, OCR, Long context, Coding | Reasoning, Coding, Function calling, Tool use, Structured output, Computer use, Web search, Long context |
| Knowledge cutoff | 2025-12 | 2026-01-04 |
| Input modalities | Text, Image, Video | Text, Image |
| Output modalities | Text | Text |
| Weights | Open weights | Open weights |
| Licence | OpenMDW License Agreement 1.1 | Apache 2.0 |
| Commercial use | Commercial use allowed | Commercial use allowed |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used Nemotron 3.5 Super VL or Muse Glimmer-30B? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.